EDITOR Kosuke Kurokawa

Energy

Feasibility of

Very Large Scale

Photovoltaic Power

Generation (VLS-PV)

Systems

fromthe

Desert

SUMMARY

Energyfrom

the

Desert

SUMMARY

EDITOR Kosuke Kurokawa

Energy

Feasibility of

Very Large Scale

Photovoltaic Power

Generation (VLS-PV)

Systems

fromthe

Desert

PART THREE: SCENARIO STUDIES AND RECOMMENDATIONS

Published by James & James (Science Publishers) Ltd

8–12 Camden High Street, London, NW1 0JH, UK

© Photovoltaic Power Systems Executive Committee of the

International Energy Agency

The moral right of the author has been asserted.

All rights reserved. No part of this book may be reproduced in any form or by

any means electronic or mechanical, including photocopying, recording or by

any information storage and retrieval system without permission in writing

from the copyright holder and the publisher.

A catalogue record for this book is available from the British Library.

Printed in Hong Kong by H&Y Printing Ltd

Cover image: Horizon Stock Images / Michael Simmons

Neither the authors nor the publisher make any warranty or representation,

expressed or implied, with respect to the information contained in this

publication, or assume any liability with respect to the use of, or damages

resulting from, this information.

Please note: in this publication a comma has been used as a decimal point, according to the

ISO standard adopted by the International Energy Agency.

CHAPTER ELEVEN: CONCLUSIONS OF PART 1 AND PART 2

Contents

Foreword vi

Preface vii

Task VIII Participants viii

COMPREHENSIVE SUMMARY

Objective 1

Background and concept of VLS-PV 1

VLS-PV case studies 1

Scenario studies 2

Understandings 2

R ecommendations 2

EXECUTIVE SUMMARY

A. Background and concept of VLS-PV 3

A.1 World energy issues 3

A.2 Environmental issues 4

A.3 An overview of photovoltaic technology 5A.3.1 Technology trends 5A.3.2 Experiences in operation and maintenance of large-scale PV systems 6A.3.3 Cost trends 6A.3.4 Added values of PV systems 7

A.4 World irradiation database 8

A.5 Concept of VLS-PV system 8A.5.1 Availability of desert area for PV technology 8A.5.2 VLS-PV concept and definition 9A.5.3 Potential of VLS-PV: advantages 10A.5.4 Synthesis in a scenario for the viability of VLS-PV development 10

B. VLS-PV case studies 12

B.1 General information 12

B.2 Preliminary case study of VLS-PV systems in world deserts 12

B.3 Case studies on the Gobi Desert from a life-cycle viewpoint 15

B.4 Case studies on the Sahara Desert 18

B.5 Case studies on the Middle East desert 19

C. Scenario studies and recommendations 21

C.1 Sustainable growth of the VLS-PV system concept 21

C.2 Possible approaches for the future 22

C.3 Financial and organizational sustainability 24

C.4 R ecommendations 25C.4.1 General understandings 25C.4.2 Recommendations on a policy level 26C.4.3 Checklist for specific stakeholders 26

PART THREE: SCENARIO STUDIES AND RECOMMENDATIONS

vi

Foreword

The International Energy Agency's Photovoltaic Power

Systems Programme, IEA PVPS, is pleased to publish this

study on very large-scale photovoltaic (VLS-PV) systems.

VLS-PV systems have been proposed on different

occasions and they may also represent a controversial theme.

The present market focus is indeed on small-scale, dis-

persed stand-alone photovoltaic power systems as well

as small and medium-sized building-integrated grid-

connected photovoltaic power systems. Both applications

have proven large potentials, of which only a very small

fraction has been realized until now. However, in the longer

term, VLS-PV systems may represent a future option for

photovoltaic applications and thereby contribute even more

to the world energy supply.

For the first time, the present study provides a detailed

analysis of all the major issues of such applications. Thanks

to the initiative of Japan, Task VIII of the IEA PVPS

Programme was designed to address these issues in a

comprehensive manner, based on latest scientific and tech-

nological developments and through close international co-

operation of experienced experts from different countries.

The result is the first concrete set of answers to some of the

main questions that have to be addressed in this context.

Experience with today’s technology is used, together with

future projections, to make quantified estimations regard-

ing the relevant technical, economic and environmental

aspects. Besides the specific issues of VLS-PV, the subject

of long-distance high-voltage transmission is also addressed.

This study includes a number of case studies in desert

areas around the world. These case studies have been

carried out in order to investigate the VLS-PV concept

under specific conditions and to identify some of the local

issues that can affect the concept.

I would like to thank the IEA PVPS Task VIII Expert

Group, under the leadership of Prof. K. Kurokawa and Dr

K. Kato, for an excellent contribution to the subject

investigated. The study provides an objective discussion base

for VLS-PV systems. This is very much in line with the

mission of the IEA PVPS Programme, aimed at objective

analysis and information in different technical and non-

technical areas of photovoltaic power systems.

I hope that this study can stimulate the long-term

discussion on the contribution of photovoltaics to the

future energy supply by providing a thorough analysis of

the subject investigated.

Stefan Nowak

Chairman, IEA PVPS Programme

vii

CHAPTER ELEVEN: CONCLUSIONS OF PART 1 AND PART 2

‘It might be a dream, but …’ has been a motive for con-

tinuing our chosen study on Very Large Scale Photovoltaic

Power Generation – VLS-PV. Now, we are confident that

this is not a dream. A desert truly does produce energy.

This report deals with one of the promising recommen-

dations for solving world energy problems in the 21st

century.

This activity first started in 1998 under the umbrella

of IEA Task VI. The new task, Task VIII: ‘Very Large Scale

PV Power Generation Utilizing Desert Areas’, was set up

for feasibility studies in 1999.

To initiate our study, a lot of imagination was required.

It was felt that dreams and imagination are really wel-

come, and that it is worth while to consider things for

future generations, our children and grandchildren.

People have to imagine their lives after 30 or 50 years,

even 100 years, since it requires a longer lead-time to

realize energy technology. In this sense, studies in terms

of VLS-PV include plant design by extending present

technologies as well as discussing basic requirements for PV

energy in the future energy-supplying structure, the social

impact on regions, and the local and global environmental

impact.

It is known that very large deserts in the world have a

large amount of energy-supplying potential. However,

unfortunately, around those deserts, the population is

Preface

generally quite limited. Then, too much power generation

by PV systems becomes worthless. However, world energy

needs will grow larger and larger towards the middle of

the 21st century. In addition, when global environmental

issues are considered, it is felt that future options are lim-

ited. These circumstances became the backbone and

motive force for VLS-PV work.

Finally, all the Task VIII experts wish to thank the IEA

PVPS Executive Committee and the participating coun-

tries of Task VIII for giving them valuable opportunities

for studies.

Prof. Kosuke Kurokawa

Editor

Operating Agent–Alternate, Task VIII

Dr Kazuhiko Kato

Operating Agent, Task VIII

PART THREE: SCENARIO STUDIES AND RECOMMENDATIONS

viii

Kazuhiko Kato, OA

New Energy and Industrial Technology Development Organization (NEDO), Japan

Kosuke Kurokawa, OA–Alternate

Tokyo University of Agriculture and Technology (TUAT), Japan

Isaburo Urabe, Secretary

Photovoltaics Power Generation Technology R esearch Association (PVTEC), Japan

David Collier

Sacramento Municipal Utility District (SMUD), USA

David Faiman

Ben-Gurion University of the Negev, Israel

Keiichi Komoto

Fuji R esearch Institute Corporation (FR IC), Japan

Jesus Garcia Martin

Iberdrola S.A., Brussels Office, Spain

Pietro Menna

General Directorate for Energy and Transport – D2, European Commission, Italy

Kenji Otani

National Institute of Advanced Industrial Science and Technology (AIST), Japan

Alfonso de Julian Palero

Iberdrola, Spain

Fabrizio Paletta

CESI SFR -ER I, Italy

Jinsoo Song

Korea Institute of Energy R esearch (KIER ), Korea

Leendert Verhoef

Verhoef Solar Energy Consultancy, the Netherlands

Peter van der Vleuten

Free Energy International bv, the Netherlands

Namjil Enebish, Observer

Department of Fuel and Energy, Ministry of Infrastructure, Mongolia

Task VIII Participants

1

Comprehensive Summary

Comprehensive summary

OBJECTIVE

The scope of this study is to examine and evaluate the

potential of very large-scale photovoltaic power genera-

tion (VLS-PV) systems (which have a capacity ranging

from several megawatts to gigawatts), by identifying the

key factors that enable VLS-PV system feasibility and

clarifying the benefits of this system’s application to

neighbouring regions, as well as the potential contribution

of system application to protection of the global environ-

ment. R enewable energy utilization in the long term also

will be clarified. Mid- and long-term scenario options for

making VLS-PV systems feasible in some given areas will

be proposed.

In this report, the feasibility and potential for VLS-PV

systems in desert areas are examined. The key factors

for the feasibility of such systems are identified and the

(macro-)economic benefits and the potential contribution

to the global environment are clarified. First the back-

ground of the concept is presented. Then six desert areas

are compared, and three of these are selected for a case

study. Finally, three scenario studies are performed to

ensure sustainability.

BACKGROUND AND CONCEPT OF VLS-PV

A very large-scale PV system is defined as a PV system

ranging from 10 MW up to several gigawatts (0,1–20 km 2

total area) consisting of one plant or an aggregation of

multiple units operating in harmony and distributed in

the same district. These systems should be studied with an

understanding of global energy scenarios, environmental

issues, socio-economic impact, PV technology develop-

ments, desert irradiation and available areas:

• All global energy scenarios project PV to become a

multi-gigawatt generation energy option in the first half

of this century.

• Environmental issues which VLS-PV systems may help

to alleviate are global warming, regional desertification

and local land degradation.

• PV technology is maturing with increasing conversion

efficiencies and decreasing prices per watt. Prices of

1,5 USD/ W are projected for 2010, which would

enable profitable investment and operation of a 100 MW

plant.

• Solar irradiation databases now contain detailed infor-

mation on irradiation in most of the world’s deserts.

• The world’s deserts are so large that covering 50 %

of them with PV would generate 18 times the world

primary energy supply of 1995.

VLS-PV CASE STUDIES

Electr icity generation costs of between 0,09 and

0,11 USD/ kWh are shown, depending mainly on annual

irradiation level (module price 2 USD/ W, interest rate 3 %,

salvage value rate 10 %, depreciation period 30 years).

These costs can come down by a factor of a half to a

quarter by 2010. Plant layouts and introduction scenarios

exist in preliminary versions. I/ O analysis shows that

25 000–30 000 man-years of local jobs for PV module

production are created per 1 km2 of VLS-PV installed.

Other findings of the three case studies (two flat-plate PV

systems and one two-axis tracking concentrator PV) are as

follows:

• The case study in the Gobi Desert describes a VLS-PV

system built of strings of 21 modules combined into

arrays of 250 kW consisting of 100 strings. Two of these

arrays are connected to an inverter of 500 kW. Two

hundred of these sets of two arrays are distributed over

an area of approximately 2 km2. Total requirements forconstruction of the plant based on local module assem-

bly are 848 485 modules, 1 700 tons of concrete for

foundations and 742 tonnes of steel for the array sup-

ports. The life-cycle CO2 emission is around 13 g-C/

kWh, due mainly to manufacturing of the modules andthe array supports.

• In the Sahara case study, several distributed generation

concepts were compared to minimize transmission costs.

A potentially attractive option is 300 dispersed plants of

5 MW PV systems, the total capacity of which is1,5 GW, located along the coast of Northern Africa,

connected to the grid by a single 1–10 km medium-

voltage line. A complete I/ O analysis was also carried

Comprehensive Summary

2

out, resulting in 2 570 induced jobs by the operation ofa 5 MW/ year PV module production facility.

• In the Negev Desert in the Middle East, a 400-sun

concentrator dish of 400 m2 was evaluated. Simulations

indicated that 16,5 % overall system efficiency is

achievable, and an economically attractive operationwith generation costs of less than 0,082 USD/ kWh is

possible.

SCENARIO STUDIES

Three sustainable scenario studies were developed showing

that sustainable local economic growth, sustainable technological–

environmental development and non-technological demonstration

and sustainable financial (stakeholder) support are possible when

a long-term perspective is developed and maintained:

• In the concept of sustainable local economic growth,

the first local PV module production facility has an

annual output of 5 MW. This local production supplies

for the construction of the local VLS-PV system. In

subsequent years, four more 5 MW module production

facilities are brought into operation, so that annually

25 MW is supplied to the local VLS-PV system. After

10–15 years, a module production facility of 50 MW

is put into operation. Every 10 years this facility is

replaced by a more modernized one. Thus after

approximately 40 years a 1,5 GW VLS-PV plant is in

operation, and the local production facility supplies for

replacement. In this way, local employment, and thus

the economy, will grow sustainably.

• To reach the point of a 1 GW system, four intermedi-

ate stages are necessary: R &D stage, pilot stage, dem-

onstration stage, and deployment (commercial) stage.

From stage to stage, the system scale will rise from

2,5 MW to 1 GW, and module and system cost will go

down by a factor of 4. Production will be shifted more

and more to the local economy. Technological issues to

be studied and solved include reliability, power control

and standards. Non-technical items include training,

environmental anti-desertification strategies, industri-

alization and investment attraction. These four stages have

a total duration of 15 years.

• To realize the final commercial stage, a view to financ-

ing distribution is developed for all of the three

previous stages, consisting of direct subsidies, soft loans,

equity, duty reduction, green certificates and tax

advantages. It is clear that direct subsidies will play an

important role in the first three stages (R &D, pilot and

demonstration). Ultimately, in the commercial stage,

enough long-term operating experience and track

record are available to attract both the soft loans and

equity for such a billion-dollar investment.

UNDERSTANDINGS

From the perspective of the global energy situation, global

warming and other environmental issues, as well as from

the case studies and scenarios, it is apparent that VLS-PV

systems can:

• contribute substantially to global energy needs

• become economically and technologically feasible

• contribute considerably to the environment

• contribute considerably to socio-economic development.

RECOM M ENDATIONS

To secure that contribution, a long-term scenario (10–15

years) perspective and consistent policy are necessary on

technological, organizational and financial issues. Action

is required now to unveil the giant potential of VLS-PV

systems in deserts. In such action, the involvement of many

actors is needed. In particular, it is recommended that, on

a policy level:

• national governments and multinational institutions

adopt VLS-PV systems in desert areas as a viable energy

generation option in global, regional and local energy

scenarios;

• the IEA-PVPS community continues Task VIII to

expand the study, refine the R &D and pilot phases,

involve participation by desert experts and financial

experts, and collect further feedback information from

existing PV plants;

• multilateral and national governments of industrialized

countries provide financing to generate feasibility studies

in many desert areas around the world and to imple-

ment the pilot and demonstration phases;

• desert-bound countries (re-)evaluate their deserts not as

potential problem areas but as vast and profitable

(future) resources for sustainable energy production, rec-

ognizing the positive influence on local economic growth,

regional anti-desertification and global warming.

3

Background and Concept of VLS-PV

EXECUTIVE SUMMARY A

Background and concept of

VLS-PV

We are in a new age beyond the 20th century, which was

the age of high-consumption society maintained by a mass

supply of fossil fuels and advances in science and technol-

ogy. But our activities in such a society will have a serious

impact on us, such as in energy security, global environ-

mental issues, population problems, etc. Therefore, it is

necessary to reconstruct a new society with new values

and new lifestyles in order to sustain our world from now

on. Finding solutions for energy and environmental issues

is essential for realizing a sustainable world, since it will

take a long time to develop energy technologies and to

recover from the destruction of the global environment.

R enewable energy such as solar, hydropower, geothermal

and biomass is expected to be the main energy resource in

future. Photovoltaic (PV) technology is one of the most

attractive options of these renewables, and many in the

world have been trying to develop PV technologies for the

long term.

In Part I, the informative introductory part of the whole

report, both global energy and environmental issues,

including the potential of renewable energy sources and

the market trends in PV technology, are reviewed as a

background for this report. General information on PV

technology, such as trends in solar cells and systems, opera-

tion and maintenance experiences, and a case study on

added values of a PV system for utilities, are summarized.

World irradiation data are also important to start a discus-

sion about the potential of VLS-PV systems. In the last

chapter of Part I, the concept of VLS-PV systems, which is

the theme of this report, is introduced.

A.1 WORLD ENERGY ISSUES

The two oil crises in the 1970s made us aware that fossil

fuels are exhaustible and triggered development of

alternative energy resources such as renewable energy.

Nevertheless, most of the primary energy still depends on

fossil fuels, and current utilization of renewables is negli-

gibly small, except for hydropower. According to the IEA

report, generally the total amount of fossil-fuel resources

in the world will not exhaust the energy supply until 2030,

although there are possibilities of a rapid increase in

energy demand, a geographical imbalance between supply

and demand, and temporal and local supply problems. There

is a forecast that the world primary energy supply in 2030

will increase to over 1,5 times as much as that in 2000, as

shown in Figure A.1.

In addition, energy demand in Asian countries will

increase much more than in O ECD countries. Even

beyond 2030, rapid growth in developing countries may

continue further, reflecting the economic gap between the

developing and the industrialized countries. In addition to

Figure A.1 World primary energy supply

by region, 1971–2030. Source: IEA

Executive Summary A

4

the long-term world energy problem, global warming is

another urgent issue because CO2 emissions are caused by

the combustion of fossil fuels (see Figure A.2). As pointed

out at Kyoto CO P-3, simple economic optimization

processes for world energy supply can no longer be

accepted to overcome global warming.

In any consideration of future energy problems, basic

conditions and tendencies may be summarized as follows:

• World energy demands will rapidly expand towards the

middle of this century due to world economic growth

and population increase.

• The sustainable prosperity of human beings can no

longer be expected if global environmental issues are

ignored.

• The share of electrical energy is rising more and more

as a secondary energy form.

• Although the need for nuclear power will increase as a

major option, difficulties in building new plants are

getting more and more notable at the same time.

• Thinking about the long lead-time for the development

of energy technology, it is urgently necessary to seek

new energy ideas applicable for the next generation.

In order to solve global energy and environmental issues,

renewable energy resources are considered to have a large

potential as well as to provide energy conservation,

carbon-lean fuels and CO2 disposal/ recovery. Among the

variety of renewable energy technologies, photovoltaic

(PV) technology is expected to play a key role in the mid-

dle of this century, as reported by Shell International

Petroleum Co. and the G8 R enewable Energy Task Force

(see Table A.1).

The world PV market as well as the world PV system

installation has been growing rapidly for the past several

years. Besides, PV industries in the USA, Europe and

Japan recently established their long-term vision of the PV

market. According to their vision, potential cumulative PV

installation will be in the hundreds of gigawatts in 2030.

A.2 ENVIRONM ENTAL ISSUES

R ecently, great concern about environmental issues, most

of which have been caused by human activities, has spread

throughout the entire world. The environmental impact of

VLS-PV systems may be divided into three categories from

a geographical viewpoint, i.e. global, regional and local

environmental issues, as shown in Figure A.3. The global

environmental issues are matters related to global changes.

R egional issues are trans-boundary environmental issues,

including atmospher ic and water pollution. Local

environmental issues are changes restricted to the local

environment that surrounds the VLS-PV installation site.

The most important phenomenon in this issue may be

desertification and land degradation. Change in micro-

climate is another local environmental issue.

Among environmental issues, global warming is one of

the most important issues because it has a large variety of

impact in various respects. According to the IPCC (Inter-

governmental Panel on Climate Change) Third Assessment

R eport, the global average surface temperature has

increased by (0,6 ± 0,2) °C since the late 19th century. It

is very likely that the period from 1990 to 2000 was the

warmest decade. Also, the global average surface tempera-

ture has been projected to increase by 1,4–5,8 °C between

1990 and 2100. The projected rate of warming is much

Table A.1 Installed global capacity estimated by G8 renewable energy task force (GW)

Coal IGCC Gas FC Bio conv. Bio IGCC Bio FC Small hydro Wind Solar PV Solar thermal Geothermal

2000 0,3 0,3 24,3 0,3 0,1 0,3 12,5 1,0 1,4 6,9

2012 14,3 14,3 34,6 15,7 15,8 18,9 90,5 31,8 9,7 17,4

2020 39,7 35,7 36,7 28,9 28,9 38,8 196,3 118,8 32,6 27,6

2030 142,5 114,0 40,5 68,0 67,5 95,8 554,6 655,8 156,4 49,5

Figure A.2 World primary energy supply and CO2 emissions,

1971–2030. Source: IEA

Figure A.3 Possible environmental issues impacted by

VLS-PV systems

5

Background and Concept of VLS-PV

greater than the observed changes during the 20th cen-

tury, and is very likely to be without precedent at least

during the last 10 000 years.

To mitigate the projected future climate change and

influences, the UN Framework Convention on Climate

Change (UNFCCC) has activated a negotiating process.

In COP-3 held in Kyoto in 1997, the Kyoto Protocol was

adopted and six greenhouse gases (GHGs) have been

designated for reduction by the first commitment period.

In November 2001, CO P-7 was held in Marrakesh,

Morocco. At this conference, the Marrakesh Accords were

adopted, and many have expressed a wish for the Kyoto

Protocol to enter into force in 2002. The finalized Kyoto

rulebook specifies how to measure emissions and reduc-

tions, the degree to which carbon dioxide absorbed by

carbon sinks can be counted towards the Kyoto targets,

how the joint implementation and emissions trading

systems will work, and the rules for ensuring compliance

with commitments. The meeting also adopted the Marra-

kesh Ministerial Declaration as input for the 10th

anniversary of the Convention’s adoption and the ‘R io+10’

World Summit for Sustainable Development (Johannes-

burg, September 2002). The Declaration emphasizes the

contribution that action on climate change can make to

sustainable development and calls for capacity building,

technology innovation and co-operation with the

biodiversity and desertification conventions.

Desertification is the degradation of land in arid, semi-

arid and dry subhumid areas. It occurs because dryland

ecosystems, which cover over one-third of the world’s land

area, are extremely vulnerable to overexploitation and

inappropriate land use. Desertification reduces the land’s

resilience to natural climate variability. Soil, vegetation,

freshwater supplies and other dryland resources tend to

be resilient. They can eventually recover from climatic

disturbances, such as drought, and even from human-

induced impacts, such as overgrazing. When land is

degraded, however, this resilience is greatly weakened. This

has both physical and socio-economic consequences.

Combating desertification is essential to ensuring the

long-term productivity of inhabited drylands. Unfortu-

nately, past efforts at combating desertification have too

often failed, and around the world the problem of land

degradation continues to worsen. R ecognizing the need

for a fresh approach, 179 governments have joined the

UNCCD as of March 2002. The UNCCD promotes

international co-operation in scientific research and

observation, and stresses the need to co-ordinate such

efforts with other related Conventions, in particular those

dealing with climate change and biological diversity.

New technologies and know-how should be developed,

transferred to affected countries, and adapted to local

circumstances. For example, photovoltaic and wind

energy may reduce the consumption of scarce fuelwood

and deforestation. These technologies, however, should also

be environmentally sound, economically viable and socially

acceptable.

VLS-PV systems will be one of the promising tech-

nologies for solving environmental problems. However, if

some projects involving environmentally safe and sound

technology are proposed, we should pay attention not only

to the operation but also to the entire life-cycle, including

production and transportation of components and inci-

dental facilities, construction and decommissioning. For

this purpose, life-cycle assessment (LCA) is a useful approach

and is becoming a general method of evaluating various

technologies. Besides the contribution to reducing gas

emissions such as CO2, projects for developing and intro-

ducing new technologies, such as the Clean Development

Mechanism (CDM), must accompany the sustainable

social and economic development of the region.

A.3 AN OVERVIEW OF PHOTOVOLTAICTECHNOLOGY

A.3.1 Technology trends

PV technology has several specific features such as solar

energy utilization technology, solid-state and static devices,

and decentralized energy systems. The long history of R &D

on solar cells has resulted in a variety of solar cells. Crystal-

line (single-crystalline, polycrystalline) silicon is the most

popular material for making solar cells. In 2001, crystal-

line Si PV modules had approximately 80 % of the market

share.

Mainly because of the lack of sufficient supply of suit-

able silicon material and because of the limited possibili-

ties for further improvements in manufacturing costs for

wafer-based silicon solar cells, much of the worldwide R &D

effort is spent on the development of thin-film solar cells.

Since extensive expertise has been gained with silicon as a

semiconductor material, the first candidates for replacing

wafer-based solar cells use silicon as an active layer. The

most popular thin-film technology today uses amorphous

silicon as the absorber layer; low-cost manufacturing tech-

niques have been designed and amorphous silicon solar

panels are the most cost-effective in the market today.

Multiple cell concepts, using a combination of amorphous

silicon and microcrystalline silicon cells (micromorph

concept), show interesting potentials for increasing solar-

cell efficiencies at relatively low cost. Another group of

thin-film silicon solar cells make use of high-temperature

deposition techniques and grow the silicon thin films on

high-temperature resistant (mostly ceramic) substrates.

Making use of lift-off and transfer techniques, silicon

layers that have been grown on silicon substrates at high

temperatures can be transferred to low-cost substrates and

the original substrate can be re-used. A different approach

using silicon thin films for enhancing the efficiency of a

silicon solar cell is the combination of crystalline silicon

wafers with amorphous silicon cells (hetero-junction cells).

Compound thin-film solar cells using material other than

silicon (CIGS, CdTe) have demonstrated their high

efficiency capability and offer a promising future for this

type of thin-film solar cells. Dye-sensitized and organic

solar cells have potential of low cost; today, their efficiencies

are still and probably will remain low, and the lifetime of

the cells is a major concern. Long-term reliability of thin-

Executive Summary A

6

film solar panels is, as with other types of semiconductors,

very much dependent on the encapsulation.

The general trend for all future design activities will be

to improve the conversion efficiency of the cell, the

simplicity, the throughput and the yield of the production

process, and the long-term reliability of the module. Look-

ing at the present status of R &D, manufacturing and

market penetration of the various technologies, it can be

expected that amorphous silicon will remain the domi-

nant thin-film technology in coming years. In particular,

the combination with microcrystalline silicon offers higher

and more stable efficiency, which is needed in many

applications. The next dominant thin-film technologies may

be polycrystalline compound solar cells, like CIS and CdTe.

Next to that, thin-film silicon cells, either on ceramic ma-

terials or via transfer techniques, may offer the best price/

performance ratio for most applications. For the longer

term, organic cell concepts may also enter the market. In

principle, all the cell concepts mentioned above have the

potential to reach and even pass the 1 USD/ W level. It can

be expected that research activities on all concepts will be

continued and that all concepts, at some time in the future,

will be commercially available. These are all major drivers

towards lower cost per unit of electricity.

A.3.2 Experiences in operation andmaintenance of large-scale PV systems

According to the facts of three projects of SMUD, opera-

tion and maintenance cost of large-scale PV system seems

to be low, less than 1 % of gross total project cost, as shown

in Table A.2.

The long-term reliability of solar cell and modules was

discussed by reviewing long-term data on field exposure

in four regions: Pacific R im, USA, Europe and Negev

Desert. In general, the performance degradation of a

crystalline Si solar-cell module ranges between 0,4 and

2,0 %/ year. In this report, performance degradation was

classified into three levels: typical (0,5 %/ year), severe

(1,0 %/ year), and worst (1,5 %/ year).

A.3.3 Cost trends

Although PV is currently at a disadvantage because of its

high cost, we believe PV has the best long-term potential

because it has the most desirable set of attributes and the

greatest potential for radical reductions in cost. Costs for

the entire system vary widely and depend on a variety of

factors, including system size, location, customer type, grid

connection and technical specifications. For example, for

building-integrated systems (BIPV), the cost of the system

will vary significantly depending on whether the system

is part of a retrofit or is integrated into a new building

structure. Another factor that has been shown to have a

significant effect on prices is the presence of a market stimu-

lation measure, which can have dramatic effects on

demand for equipment in the target sector. The installation

of PV systems for grid-connected applications is increas-

ing year by year, while the grid-connected market must

still depend upon government incentive programmes at

present. The installed cost of grid-connected systems also

varies widely in price. Figure A.4 shows the trends of PV

system and module prices in some countries. Although, in

more recent years, this shows a slight increase in some

markets due to high demand, there appears to be a contin-

ued downward trend.

We need to accelerate that trend. One way to do that is

to step up the scale of the typical PV plant. The largest

plant has a capacity approaching 100 MW/ year. It would

take such a plant, running flat out, 100 years to produce

enough equipment to match the power-generating capac-

ity of one medium-sized combined-cycle gas turbine power

plant. We believe there may be significant economies of

scale to be reaped as we move up to 50–100 MW plants.

Another path towards radically lower costs is technology

step change. The technology in use today is based on

crystalline silicon. This is an inherently material-intensive

technology. It requires batch production methods, and

is now relatively mature. The great hope for the future

lies with thin-film technologies, which are much less

material-intensive and suitable for continuous production

Table A.2 Actual operation and maintenance costs

Project name Total project cost Actual Actual Actual (USD)

(USD) (USD) (USD/kW) / Total project cost

Solarex Residential (329 kW) 2 050 723 6 502 19,76 0,32 %

Sacramento Metropolitan Airport

Solarport (128 kW) 1 324 122 7 500 58,59 0,68 %

Rancho Seco PV-3 Ground-Mounted

Substation System (214 kW) 2 580 008 4 167 19,47 0,25 %

Figure A.4 PV module and system price trends in some

countries

7

Background and Concept of VLS-PV

processes. They offer the potential to shift on to a lower

and steeper learning curve. However, we need to be a little

cautious about predicting when thin film will start to

realize its commercial potential. Both of these routes –

stepping up the scale, and backing the new technology –

carry large risks, both technical and commercial. Taking

bold steps will require a great deal of confidence in the

rapid emergence of a mass market.

Today’s cost of single-crystal and polycrystalline silicon

modules (although proprietary) is such that the present

factory price of 4 USD/ W includes all costs, as well as

marketing and management overheads, for that product

line. N ote that module prices for single-crystal and

polycrystalline silicon have been essentially stable, between

3,75 and 4,15 USD/ W, for nearly 10 years, while manu-

facturing costs have been reduced by over 50 %. Based on

the studies cited and on further analysis, it seems likely

that fully loaded manufacturing costs for a 100 MW

single-crystal silicon module will be 1,40 USD/ W. This

would permit a profitable price of 2,33 USD/ W. Prices

at this level are likely to be required in order to open up

the massive grid-connected and building-integrated

markets. The cast-ingot polycrystalline option continues

to have module costs slightly lower than those of the

single-crystal, allowing this option to offer profitable prices

at the 2 USD/ W level. Thin films and concentrators could

have manufacturing costs that will allow profitable prices

of 1,25 USD/ W. In this forecast, the PV market contin-

ues to grow at 15–25 %. However, in order for this fore-

cast to become a reality, several major events need to

occur. First, the existing subsidized grid-connected

programmes in countries such as Japan, Germany, the

Netherlands and the USA need to stimulate the installa-

tion of quality, reliable systems in sufficient quantities to

stimulate investment in large-volume module manufac-

turing plants. Secondly, continuing decrease in the price

of modules must be realized. There must be profitable

modules below 2 USD/ W by the year 2005 and even

lower prices, approaching 1,5 USD/ W, must occur by

2010. Thirdly, in the transitional phase towards competi-

tive prices, marketing and financing schemes have to be

introduced more widely which will allow the customers

to opt for this solution in spite of costs.

A.3.4 Added values of PV systems

PV technology has unique characteristics different from

those of conventional energy technologies, and additional

values are hidden in PV systems besides their main func-

tion, which is, of course, power generation. Table A.3 shows

a summary of non-energy benefits that can add value to

PV systems. Nowadays, many people are becoming aware

of the additional benefits offered by PV systems. Unfortu-

nately, this awareness does not contribute to the effective

promotion of PV systems since current added values are

not quantitative but qualitative. Thus research activities on

quantitative analysis of this issue should be continued.

There is a case study on the added values of a PV system

for SMUD. The utility benefits evaluated are as follows.

• Energy: avoided marginal cost of system-wide energy

production.

• Capacity: avoided marginal cost of system-wide gen-

eration capacity.

• Distribution: distribution capacity investment deferral.

• Sub-transmission: sub-transmission capacity investment

deferral.

• Bulk transmission: transmission capacity investment

deferral.

• Losses: reduction in electricity losses.

• R EPI: renewable energy production incentive.

• Externalities: value of reduced fossil emissions.

• Green pricing: voluntary monthly contributions from

PV pioneers.

• Fuel price risk mitigation: value of reducing risk from

uncertain gas price projections.

Table A.3 Summary of non-energy benefits that can add value to PV systems

Category Potential values

Electrical kWh generated; kW capacity value; peak generation and load matching value; reduction in demand for ut ility

electricity; power in t imes of emergency; grid support for rural lines; reduced transmission anddistribution losses;

improved grid reliability and resilience; voltage control; smoothing load f luctuations; f iltering harmonics and

reactive power compensation.

Environmental Signif icant net energy generator over its lifet ime; reduced air emissions of part iculates, heavy metals,

CO2, NO

x, SO

x, result ing in lower greenhouse gases; reduced acid rain and lower smog levels; reduced

power station land and water use; reduced impact of urban development; reduced tree clearing for fuel;

reduced nuclear safety risks.

Architectural Substitute building component; mult i-function potential for insulation, water proofing, f ire protection,

w ind protection, acoustic control, daylighting, shading, thermal collection and dissipation; aesthetic appeal

through colour, transparency, non-reflective surfaces; reduced embodied energy of the building;reflection of

electromagnetic waves; reduced building maintenance and roof replacements.

Socio-economic New industries, products and markets; local employment for installat ion and servicing; local choice, resource use

and control; potential for solar breeders; short construction lead-t imes; modularity improves demand matching;

resource diversif ication; reduced fuel imports; reduced price volatility; deferment of large capital outlays for

central generating plant or transmission and distribution line upgrades; urban renewal; rural development; lower

externalit ies (environmental impact, social dislocation, infrastructure requirements) than fossil fuels and nuclear;

reduced fuel transport costs and pollut ion from fossil-fuel use in rural areas; reduced risks of nuclear accidents;

symbol for sustainable development and associated education; potential for international co-operation,

collaboration and long-term aid to developing countries.

Executive Summary A

8

• Service revenues (economic development): net service

revenues from local PV manufacturing plant (result of

economic development efforts).

Table A.4 is an estimation result of utility benefits of fixed

PV systems.

A.4 WORLD IRRADIATION DATABASE

Irradiation data are important to start a discussion about

the potential of VLS-PV systems. The Japan Weather Asso-

ciation (JWA) collected irradiation and air temperature

data during 1989 and 1991 from every meteorological

organization in the world. Data items are monthly means

of global irradiation, monthly means of ambient air tem-

perature, and monthly means of snow depth. The data were

collected from 150 countries, and data from 1 601 sites

throughout the world are available. Monthly global irra-

diation was estimated from monthly sunshine duration

where there were no irradiation data.

The Negev R adiation Survey, which was established in

the 1980s by the Israel Ministry of National Infrastruc-

tures, monitors the following meteorological parameters

at nine stations in the Negev Desert: normal direct beam

irradiance, global horizontal irradiance, ambient tem-

perature, humidity ratio, wind speed, and wind direction.

The data are available from the Ben-Gurion National

Solar Energy Centre, in the form of a CD-ROM, which

contains a set of Typical Meteorological Year (TMY) files

(updated every three years) together with all previous years

of actual data for each site.

Besides these, there are a variety of worldwide databases

of solar energy resources. Table A.5 shows the name and

website address for some of these.

A.5 CONCEPT OF VLS-PV SYSTEM

A.5.1 Availability of desert area forPV technology

Solar energy is low-density energy by nature. To utilize it

on a large scale, a massive land area is necessary. However,

one-third of the land surface of the Earth is covered by

very dry deserts, as shown in Figure A.5. High-level inso-

lation and large spaces exist. It is estimated that if a very

small part of these areas, say 4 %, were used for the instal-

lation of PV systems, the annual energy production would

equal world energy consumption.

A rough estimation was made to examine the potential

of desert under the assumption of a 50 % space factor for

installing PV modules on the desert surface as the first

evaluation. The total electricity production becomes

Table A.4 Estimation result of ut ility benefits of f ixed PV systems (USD/kW, 1996)

Benefits Bulk transmission Sub-transmission Primary voltage Secondary voltage

Service revenues 708 708 708 708

REPI 221 221 221 221

Externalit ies 324 327 338 340

Fuel price risk 192 194 200 201

Green pricing 0 0 44 44

Distribution 0 0 0 117

Sub-transmission 0 0 39 39

Bulk transmission 0 15 16 16

Generation capacity 296 300 314 315

Energy 768 775 800 805

Total 2 509 2 540 2 680 2 806

Table A.5 Examples of world irradiation database

Name Website address

Ground observation

1. Negev Radiation Survey http://www.bgu.ac.il/solar

2. WRDC solar radiation and radiation balance data http://wrdc-mgo.nrel.gov/

3. BSRN: Baseline Surface Radiation Network http://bsrn.ethz.ch/

4. NOAA NCDC GLOBALSOD http://www.ncdc.noaa.gov/

5. METEONORM 2000 (commercial product) http://www.meteotest.ch/

Satellite-derived data

6. SeaWiFS surface solar irradiance http://www.giss.nasa.gov/data/seaw ifs/

7. LaRC Surface Solar Energy dataset (SSE) http://eosweb.larc.nasa.gov/sse/

8. ISCCP datasets http://isccp.giss.nasa.gov/isccp.html

Figure A.5 World deserts (unit: 104 km2)

9

Background and Concept of VLS-PV

1 942,3 × 103 TWh(= 6,992 × 1021 J = 1,67 × 105 Mtoe),

which means a level almost 18 times as much as the world

primary energy supply, 9 245 Mtoe (107,5 × 103 TWh =

3,871 × 1020 J) in 1995. These are quite hypothetical

values, ignoring the presence of loads near these deserts.

However, at least these indicate high potential as primary

resources for developing districts located in such solar-

energy-rich regions.

Figure A.6 also shows that the Gobi Desert area be-

tween the western part of China and Mongolia can gener-

ate as much electricity as the present world primary

energy supply. In Figure A.7, an image of a VLS-PV

system in a desert area is shown.

A.5.2 VLS-PV concept and definition

Presently, three approaches are under consideration to en-

courage the spread and use of PV systems.

(a) Establish small-scale PV systems that are independ-

ent of each other. There are two scales for such systems:

installing stand-alone, several hundred-watt PV systems for

private dwellings; and installing 2–10 kW systems on the

roofs of dwellings as well as 10–100 kW systems on office

buildings and schools. Both methods are already being used.

The former is used to furnish electrical power in develop-

ing countries, the so-called SHS (solar home system), and

the latter is used in Western countries and in Japan. This

Figure A.6 Solar pyramid

Figure A.7 Image of a VLS-PV system in a desert area

Executive Summary A

10

seems to be used extensively in areas of short- and me-

dium-term importance.

(b) Establish 100–1 000 kW mid-scale PV systems on

unused land on the outskirts of urban areas. The PVPS/

Task VI studied PV plants for this scale of power genera-

tion. Systems of this scale are in practical use in about a

dozen sites in the world at the moment, but are expected

to increase rapidly in the early 21st century. This category

can be extended up to multi-megawatt size.

(c) Establish PV systems larger than 10 MW on vast,

barren, unused lands that enjoy extensive exposure to sun-

light. In such areas, a total of even more than 1 GW of PV

system aggregation can be easily realized. This approach

makes it possible to install quickly a large number of PV

systems. When the cost of generated electrical power is

lowered to a certain level in the future, many more PV

systems will be installed. This may lead to a drastically lower

cost of electricity, creating a positive cycle between cost

and consumption. In addition, this may become one of the

solutions to future energy and environmental problems

across the globe, and ample discussion of this possibility is

believed to be worthwhile.

The third category corresponds to very large-scale PV (VLS-

PV) systems. The definition of VLS-PV may be summa-

rized as follows:

• The size of a VLS-PV system may range from 10 MW

to one or a few gigawatts, consisting of one plant, or an

aggregation of many units that are distributed in the

same district and operate in harmony with each other.

• The amount of electricity generated by VLS-PV

systems can be considered significant for people in the

district, in the nation or in the region.

• VLS-PV systems can be classified according to the fol-

lowing concepts, based on their locations:

• land based (arid to semi-arid, deserts)

• other concept (water-based, lakes, coastal, interna-

tional waters)

• locality options (D.C.: lower, middle, higher income;

large or small countries; OECD countries).

Although VLS-PV systems include water-based options,

in principle, many different types of discussions are

required on this matter. It is not neglected but it is treated

as a future possibility outside the major efforts of this

study.

A.5.3 Potential of VLS-PV: advantages

The advantages of VLS-PV systems are summarized as

follows:

• It is very easy to find land around deserts appropriate

for large energy production by PV systems.

• Deserts and semi-arid lands are, normally, high-

insolation areas.

• The estimated potentials of such areas can easily supply

world energy needs in the middle of the 21st century.

• When large-capacity PV installations are constructed,

step-by-step development is possible through utilizing

the modularity of PV systems. According to regional

energy needs, plant capacity can be increased gradually.

This is an easier approach for developing areas.

• Even very large installations are quickly attainable to

meet existing energy needs.

• R emarkable contributions to the global environment

can be expected.

• When a VLS-PV system is introduced to a certain re-

gion, other types of positive socio-economic impact may

be induced, such as technology transfer to regional PV

industries, new employment and economic growth.

• The VLS-PV approach is expected to have a major,

drastic influence on the ‘chicken-and-egg’ cycle in the

future PV market. If this does not happen, the distance

to VLS-PV systems may become a little far.

These advantages make it a very attractive option, and

worthy of discussion regarding global energy in the 21st

century. The image of this concept is illustrated by Figure

A.8.

A.5.4 Synthesis in a scenario for the viabilityof VLS-PV development

Basic case studies were reported concerning regional

energy supply by VLS-PV systems in desert areas, where

solar energy is abundant. According to this report, the

following scenario is suggested to reach a state of large-

scale PV introduction. First, the bulk systems that have

been installed individually in some locations would be

interconnected with each other by a power network. Then

they would be incorporated with regional electricity

demand growth. Finally, such a district would become a

large power source. This scenario is summarized in the

following stages according to Figure A.9:

1 A stand-alone, bulk system is introduced to supply elec-

tricity for surrounding villages or anti-desertification

facilities in the vicinity of deserts.

2 R emote, isolated networks germinate. Plural systems are

connected by a regional grid. This contributes to load

levelling and the improvement of power fluctuation.

Figure A.8 Global network image

11

Background and Concept of VLS-PV

3 The regional network is connected to a primary trans-

mission line. Generated energy can be supplied to a load

centre and industrial zone. Total use combined with other

power sources and storage becomes important for match-

ing to the demand pattern and the improvement of the

capacity factor of the transmission line. Furthermore,

around the time stage 3 is reached, in the case of a

south-to-north inter-tie, seasonal differences between

Figure A.9 Very large-scale PV system deployment scenario

demand and supply can be adjusted. An east-to-west tie

can shift peak hours.

4 Finally, a global network is developed. Most of the

energy consumed by human beings can be supplied

through solar energy. For this last stage, a breakthrough

in advanced energy transportation will be expected on

a long-term basis, such as superconducting cables, FACTS

(flexible A.C. transmission system), or chemical media.

Executive Summary B

12

EXECUTIVE SUMMARY B

VLS-PV case studies

VLS-PV systems are expected to generate a variety

of advantages and will be one of the most attractive

technologies for our future, particularly from the view-

points of energy and the environment. However, realizing

VLS-PV systems may constitute a long-term project, the

nature of which we have not yet experienced. Therefore,

the capability of VLS-PV systems and the configuration of

each component must be assessed while taking into

account site conditions, regional electricity demand,

system performance, transmission technology or other

alternative options, and concurrent use with other energy

resources.

In Part II, some case studies on VLS-PV systems for

selected regions were undertaken to employ the concepts

of VLS-PV systems, after surveying general information

concerning candidate sites. As case studies, first, introduc-

tory analyses on the generation costs of VLS-PV systems

in world deserts were carried out. Secondly, energy

payback time, CO2 emissions and generation costs for

VLS-PV systems in the Gobi Desert in China were evalu-

ated in detail from a life-cycle point of view. Next, by

assuming the Sahara Desert as the site, a network concept

for a VLS-PV system was discussed and an estimation of

the socio-economic impacts of technology transfer were

analysed. Finally, using a simulation model for the Negev

Desert located in the Middle East, fixed modules, one-axis

tracking modules, two-axis tracking modules and a con-

centration system were summarized as expected technolo-

gies for VLS-PV systems.

B.1 GENERAL INFORM ATION

The deserts that are the most promising sites for VLS-PV

systems cover one-third of the land surface. The distribu-

tion is shown in Figure B.1. Because the degree of the

impact of VLS-PV systems would depend upon the

various conditions of particular regions, major indicators

concerning major regions/ countries with deserts were

investigated. The selected regions/ countries were China,

India, the Middle East, North Africa, Mexico and Australia.

Although the economic state of these countries (except

for Australia) is lower than the world average, it has been

growing every year. Particularly, the growth in China and

India is remarkable. With economic growth, energy con-

sumption is increasing, most of which has depended upon

fossil fuels for the commercial and industrial sectors. The

trends for electricity were almost the same. In China, In-

dia and Australia, more than 75 % of the total electricity is

generated by coal, while in the Middle East, North Africa

and Mexico, oil and gas are the main resources for gener-

ating electricity. However, the candidate site is not limited

to only one desert. That is to say, it may be concluded that

all those deserts have the possibility to be candidate sites

for VLS-PV systems.

As the methodologies for case studies, we focused on

life-cycle assessment and I/ O analysis. The former is a method

for making environmental decisions, and is becoming in-

creasingly popular in environmental policies. The latter is

a method for estimating economic effects using an inter-

industry table. These are effective methods to evaluate the

impacts of VLS-PV systems. The former is applied in a

case study on the Gobi Desert, and the latter is applied in

a case study on the Sahara Desert focusing on socio-

economic impact of VLS-PV systems.

B.2 PRELIM INARY CASE STUDY OF VLS-PVSYSTEM S IN WORLD DESERTS

To make a rough sketch of VLS-PV systems in desert areas

and to investigate their economic feasibility, a preliminary

case study was carried out. It was assumed that VLS-PV

systems each with a 1 GW capacity would be installed in

the six major deserts of the world, as shown in Figure B.2.

It was supposed that a 1 GW VLS-PV system, which is

an aggregation of ten 100 MW PV systems with flat-plate

fixed array structures, would be installed in each desert.

Figure B.3 shows conceptual images of a 1 GW VLS-PV

system. Assuming that the power output from a VLS-PV

system would be transmitted to a given load centre,

construction of 110 kV transmission lines would be taken

into account. Though the transmission lines depend upon

distance from the load centre to the VLS-PV system, a

distance of 100 km was employed for all deserts to avoid

complicated evaluation in this study.

13

VLS-PV Case Studies

Figure B.1 Aridity zones of the world.

Source: World Atlas of Desertif ication (UNEP, 1992)

Figure B.2 The six deserts used in this case study: (1) desert name, (2) reference point, (3) area, (4) annual average

ambient temperature, (5) annual horizontal global irradiation

Figure B.3 Conceptual image of a 1 GW VLS-PV system

Executive Summary B

14

To calculate the annual power generation of the

VLS-PV systems, cell temperature factors, load match-

ing factors, efficiency deviation factors and inverter mis-

match factors were taken into account in calculating the

performance ratio (PR ) for each installation site. The

annual average in-plane irradiation was estimated from

the annual global horizontal irradiation using a method

for separating into direct and scattered radiation known

as the Liu–Jordan model.

Initial costs, consisting of system component costs,

transportation costs of the system components, system

construction costs, and annual operation and maintenance

(O&M) costs, were calculated in order to estimate the

generation costs of VLS-PV systems installed in the six

world deserts. No land cost was taken into account in this

study, and cost data in Japanese yen were converted to US

dollars at the (recent) exchange rate of 120 JPY/ USD.

Based on estimates of the initial cost and the annual

O&M cost, the total annual costs of 100 MW PV systems

installed in the six deserts were calculated assuming an

annual interest rate of 3 %, a salvage value rate of 10 %, a

depreciation period of 30 years, and an annual property

tax rate of 1,4 %. An annual overhead expense of 5 % of

annual O&M costs was also taken into account. The gen-

eration costs are shown in Table B.1. The lowest generation

costs were estimated when the array tilt angle was 20°

independent of PV module price, except for the case of

the Gobi Desert, where a tilt angle of 30° had the cheapest

generation cost. The generation costs at a PV module price

of 1 USD/ W, which range from 5,2 to 8,4 US cents/ kWh,

are roughly one-third as great as those at a PV module

price of 4 USD/ W.

Figure B.4 represents the best estimates of generation

costs for each desert as a function of annual global hori-

Table B.1 Generation cost of 100 MW PV system (US cents/kWh)

Sahara Negev Thar Sonoran Great Sandy Gobi

Nema Ouarzazate

Module price =

4 USD/W

Tilt angle =

10º 14,8 18,5 20,6 17,8 18,4 19,1 19,1

Tilt angle =

20º 14,7 17,9 20,0 17,2 17,9 18,8 18,1

Tilt angle =

30º 15,1 17,9 20,4 17,3 18,0 19,4 17,6

Tilt angle =

40º 16,1 18,3 21,3 17,9 18,7 20,5 17,6

Module price =

3 USD/W

Tilt angle =

10º 11,6 14,5 16,6 14,0 14,5 15,5 15,0

Tilt angle =

20º 11,5 14,0 16,1 13,6 14,1 15,4 14,2

Tilt angle =

30º 11,8 14,0 16,5 13,7 14,2 15,9 13,8

Tilt angle =

40º 12,7 14,4 17,3 14,3 14,8 16,8 13,8

Module price =

2 USD/W

Tilt angle =

10º 8,4 10,5 12,6 10,4 10,6 12,0 10,8

Tilt angle =

20º 8,4 10,2 12,3 10,0 10,3 11,9 10,3

Tilt angle =

30º 8,6 10,2 12,7 10,2 10,5 12,4 10,0

Tilt angle =

40º 9,3 10,6 13,4 10,7 11,0 13,2 10,0

Module price =

1 USD/W

Tilt angle =

10º 5,2 6,5 8,6 6,6 6,7 8,5 6,7

Tilt angle =

20º 5,2 6,3 8,4 6,4 6,5 8,4 6,3

Tilt angle =

30º 5,4 6,4 8,8 6,6 6,7 8,8 6,2

Tilt angle =

40º 5,8 6,7 9,5 7,1 7,2 9,5 6,2

Figure B.4 Best estimates of generation cost for each desert as a function of annual global horizontal irradiation

15

VLS-PV Case Studies

zontal irradiation. With the exception of the Negev Desert

and the Great Sandy Desert, in which the generation costs

are relatively higher because of high wages, the level of

generation costs of a 100 MW PV system are roughly

the same. Though the generation costs decrease gently

according to the increase in annual irradiation, even the

generation cost for the Gobi Desert, which has much less

annual irradiation than the Sahara Desert, is on a level

with that of the Sahara. Electricity from VLS-PV systems

in these deserts would not be so cheap when the PV mod-

ule price is expensive (such as at 4 USD/ W), but the cost

of the electricity will become economic even with the

proven system technologies employed in this study when

the PV module price is reduced to a level of 1 USD/ W.

The current market price of PV modules is not low enough

for the realization of VLS-PV systems, but it is expected

that PV module prices will decrease rapidly with the growth

of the PV market. Therefore, VLS-PV systems in desert

areas will be economically feasible in the near future.

B.3 CASE STUDIES ON THE GOBI DESERTFROM A LIFE-CYCLE VIEWPOINT

As shown previously, the introduction of VLS-PV systems

in desert areas seems to be attractive from an economic

point of view when PV modules are produced at a low

price level, even though existing PV system technology is

adopted. But we must pay attention not only to the eco-

nomic aspect but also to the energy and environmental

aspects, since PV systems consume a lot of energy at their

production stage and therefore emit carbon dioxide (CO2)

indirectly, as a result. Therefore, the feasibility of VLS-PV

systems was evaluated in depth from a life-cycle view-

point by means of life-cycle analysis (LCA).

The Gobi Desert was chosen as the installation site of

VLS-PV systems for LCA in this study. This desert, which

lies in both China and Mongolia, is around 1,3 × 106 km2

in size and is located between 40°N and 45°N. Installation

of the VLS-PV systems in the Gobi Desert has some ad-

vantages: it is a stone desert rather than sand, and a utility

grid exists relatively close to the desert. In this study, it was

assumed that the 100 MW VLS-PV system would be in-

stalled in the Gobi Desert on the Chinese side.

To execute LCA, a life-cycle framework of the VLS-PV

system has to be prepared. Figure B.5 gives an image of

the life-cycle framework of the VLS-PV system in this

study. It was supposed that array support structures, trans-

mission towers and foundations for the array support struc-

tures and the transmission towers would be produced in

China and that other system components would be manu-

factured in Japan. All the components are transported to

some installation site near Hoh-hot in the Gobi Desert by

marine and/ or land transport. Land transport is also taken

into consideration. Land cost is not considered, but land

preparation was considered here.

In this study, a south-facing fixed flat array structure

was employed and the array tilt angle was given as a

variable parameter (10°, 20°, 30°, 40°). Both PV module

price and inverter price were also dealt with as variable

parameters. System performance ratio (PR ) was assumed

to be 0,78 by consideration of operation temperature, cell

temperature factor, load matching factor, efficiency

deviation factor and inverter mismatch factor (= 0,90). It

should be noted that the efficiency deviation factor

involves long-term performance degradation (0,5 %/ year)

as well as short-term surface degradation by soil (= 0,95).

The number of PV modules in a string was taken to be

21 by consideration of the Voc

of the PV module and the

D.C. voltage of the inverter. Then the rated output from

one string is 2,52 kW. Accordingly a 250 kW PV array

requires 100 strings. Two of these 250 kW PV arrays

located north and south in parallel form the 500 kW

system with a 500 kW inverter and a 6,6 kV/ 500 V trans-

former.

Figure B.6 illustrates an example of the field layout for

such a 100 MW VLS-PV system with a 30° tilt angle. It

was assumed that array supports were made of zinc-plated

stainless steel (SS400), and the thickness of several types of

steel material were chosen according to stress analysis

assuming that the wind velocity is 42 m/ s, based upon the

design standard of structural steel by the Japanese Society

of Architecture. A cubic foundation made of concrete was

used. Its dimension was decided in accordance with the

design standard of support structures for power transmission

Figure B.5 Life-cycle framework of the case studyFigure B.6 Example of f ield layout for 100 MW VLS-PV

system (30° t ilt angle) (unit: m)

Executive Summary B

16

by the Institute of Electrical Engineering in Japan. Taking

into account Japanese experience in civil engineering

and the local labour situation in China, local labour re-

quirement was also estimated for system construction such

as PV module installation, array support installation, pro-

duction and installation of foundations, cable installation,

and installation of common apparatus. Table B.2 shows a

summary of requirements to construct a 100 MW

VLS-PV system in the Gobi Desert in China.

Labour cost for the operation was estimated based on

the assumption of a 100 MW VLS-PV system that was in

service 24 hours a day by nine persons working in shifts.

The annual labour cost for electrical engineering was

assumed for these operators. Maintenance cost was also

calculated based on actual results of a PVUSA project; that

is, the cost of repair parts was 0,084 %/ year of the total

construction cost and labour for maintenance was one

person per year.

Figure B.7 shows the results of the generation cost of the

100 MW PV system. Differences in the annual cost due to

PV module price resulted in differences in these genera-

tion costs. Though the least annual cost was obtained at the

least array tilt angle for any tilt angle, the minimum

generation cost appeared at around 30° due to the in-

crease in annual power generation as the array tilt angle

increased. The generation cost stays at a high level, around

18 US cents/ kWh (just less than the current consumer price

for residential sectors in Japan), when the PV module price

is 4 USD/ W. But it decreases remarkably to less than

7,0 US cents/ kWh (close to the current electricity tariff

in China) if the PV module price goes down to 1 USD/ W.

Figure B.8 represents the results of total primary

energy requirement and energy payback time (EPT). EPT

was estimated assuming that electricity from the PV

system would replace utility power in China where

recent conversion efficiency is around 33 %. As shown in

these figures, the best EPT was obtained at 20° array tilt

angle, and BOS components made of steel and concrete

(such as array supports, transmission lines, foundations

and troughs) contributed much to both energy require-

ments because these materials consume a great deal of

energy to produce in China. Transportation also uses

a certain amount of energy. Nevertheless, EPT resulted

in a very low level. This suggests that the total energy

requirement throughout the life-cycle of the PV system

(considering production and transportation of system

components, system construction, operation and mainte-

nance) can be recovered in a short period much less than

its lifetime. Therefore VLS-PV is useful for energy

resource savings.

Figure B.9 shows the results of life-cycle CO2 emissions

and life-cycle CO2 emission rate of the 100 MW PV sys-

tem, assuming 30-year operation periods. Discussion of

these results is the same as for the total primary energy

requirement and the EPT. Considering the CO2 emission

rate of existing coal-fired power plants, about 300g-C/

kWh, the life-cycle CO2 emission rate of a 100 MW PV

system is much lower. So the VLS-PV system in desert

Table B.2 Summary of total requirements for a 100 MW VLS-PV system in the Gobi Desert

Item Unit 10º 20º 30º 40º

Material requirement

PV modulea piece 848 485 848 485 848 485 848 485

Array support structurea ton 8 291 8 606 9 658 10 763

Foundationa m3 46 487 46 487 69 391 98 801

Cablea 600 V CV 2 mm2 (single core) km 1 178 1 364 1 434 1 499

600 V CV 8 mm2 (double core) km 173 173 173 173

600 V CV 60 mm2 (single core) km 67 – – –

600 V CV 100 mm2 (single core) km – 88 106 122

6,6 kV CV-T 22 mm2 km 21 27 33 37

6,6 kV CV 200 mm2 (single core) km 38 38 38 38

110 kV CV 150 mm2 (single core) km 11 13 14 15

Trougha m3 35 235 37 450 39 406 41 037

Common apparatus

Inverter (w ith transformer)a set 202 202 202 202

6,6 kV capacitorb set 202 202 202 202

6,6 kV GIS set 4 4 4 4

110 kV/6.6kV transformerb set 5 5 5 5

110 kV GIS set 4 4 4 4

2,4 MVA capacitor set 1 1 1 1

Common power board set 1 1 1 1

Transportation

Heavy oil consumption ton 145 145 147 148

Diesel oil consumption kl 6 101 6 261 7 941 10 031

Transmission

Cable 110 kV TACSR 410 mm2 km 134 134 134 134

A.C. 70 mm2 km 16,7 16,7 16,7 16,7

Pylon (steel) ton 742 742 742 742

Foundation ton 1 715 1 715 1 715 1 715

Construction

Diesel oil consumption kl 128 170 207 238

Labour requirement man-year 2 711 2 752 2 831 2 911

a99 % of construction yield is considered.bSpare sets are included.

17

VLS-PV Case Studies

Figure B.7 Generation cost of a 100 MW PV system

Figure B.8 Total primary energy (TPE) requirement and EPT of a 100 MW PV system

Figure B.9 Life-cycle CO2 emissions and life-cycle CO

2 emission rate of a 100 MW PV

system

Executive Summary B

18

areas is a very effective energy technology for preventing

global warming.

B.4 CASE STUDIES ON THE SAHARA DESERT

It is necessary to build a concept for supplying the elec-

tricity that VLS-PV generates, and the supply of such

electricity may contribute to regional development.

Additionally, VLS-PV may induce some economic impacts,

such as an employment effect. A network concept for

introducing VLS-PV in the Sahara Desert and the

expected socio-economic impact of the technology

transfer of PV module fabrication were discussed.

A transmission system devoted to the exploitation of

remote energy resources has to be designed to minimize

transmission costs, while respecting reliability and

environmental requirements. Transmission costs depend on

the transmission distances and the hours of yearly utiliza-

tion. From these viewpoints, the transmission costs were

analysed for three cases: (1) a 1,5 GW centralized PV power

plant plus 300–900 km of transmission line; (2) 1,5 GW

produced by 30 PV plants of 50 MW; and (3) 1,5 GW

produced by 300 PV plants of 5 MW. The third case

produced the most attractive results. Each of these plants

should cover an area of approximately 10 ha (0,1 km2),

and the power would be typically delivered through single

A.C. MV lines (for example, 20 kV). In this case, the PV

plants would be distributed within the coastal strip of North

African countries, placed less than 10 km from the HV/

MV substations and the distribution networks that feed

the loads. When assuming 5 km as a transmission distance,

the transmission cost would range from 3,7 to 6,2 USD/

MWh, depending upon yearly utilization of VLS-PV, as

shown in Figure B.10.

Additionally, having a variety of technology transfers is

very important for the sustainable expansion of VLS-PV,

and the transfer of PV manufacturing facilities may bring

about the stimulation of various economic activities as well

as the establishment of a local PV industry. To grasp such

impacts quantitatively, a technical analysis of an industrial

initiative in the photovoltaic sector and evaluation of the

socio-economic impacts of the PV demand in terms of

gross domestic production and job creation were carried

out by the I/ O analysis method. When assuming the trans-

fer of a facility with the capacity of 5 MW/ year, it has

been made clear that the local availability of cells (Case L)

brings very different returns on investment in the local

economy, and the induced production increases from

1,4 times to 3,5 times the expenditure, as shown in Figure

B.11 and Table B.3. The induced job creation involved

Figure B.10 Case of small PV systems along w ith

netw ork

Figure B.11 Necessary expenditure and induced production (MUSD)

19

VLS-PV Case Studies

2 570 employees (Case L), while on the other hand, in the

case of not assuming the availability of local cell produc-

tion (Case I), the induced job creation involved 489

employees.

Besides the economic effects of manufacturing, the

availability of PV systems contributes to support social and

economic development of the region that is more

environmentally sound. To perform technology transfer

effectively, clear knowledge of the needs of, as well as the

potential benefits to, the local population is required.

However, it is expected in the future that VLS-PV will

be reinforced year by year with operating PV module

facilities in the region, and that the electricity generated

by VLS-PV will be able to supply global areas through the

Mediterranean Network. As a result, an electricity-for-

technology exchange scheme may be initially set up.

B.5 CASE STUDIES ON THEM IDDLE EAST DESERT

There are certain types of PV systems considered for

constructing VLS-PV systems. In addition to the fixed-

module VLS-PV, the use of sun-tracking non-concentrator

and concentrator PV systems is expected. The relative

performances – which involved static PV modules

(oriented facing south, with tilt angle equalling geographic

latitude), one-axis tracking modules (having a horizontal

axis in the N–S direction), two-axis tracking modules, and

a 400× point-focus concentrator PV system – are addressed

and the potential economic benefits of employing highly

concentrated solar radiation as an energy source for PV

cells were considered. As a practical site, the Negev Desert

was chosen, because the Negev is located at a truly repre-

sentative ‘point’ among the principal deserts of the Middle

East and has produced a wealth of documented detailed

meteorological data.

Computer simulations were carried out, using a typical

meteorological year (TMY) dataset for a specific Negev

site, Sede Boqer. At Sede Boqer, a 25 m diameter multi-

purpose solar concentrator operating at a solar concentra-

tion of 400×, as shown in Figure B.12, already exists.

Simulations have indicated that, in a typical Middle

Eastern desert, 8,5 % total system efficiency from a

conventional static PV system, 10,7 % effective system

efficiency for a one-axis sun-tracking system, and 11,8 %

effective system efficiency for a two-axis tracking system

are to be expected, where all three system types employ

identical, polycrystalline Si PV modules. However, using

a dense array of Si concentrator PV cells in a 400-sun

point-focus system, the simulations have indicated that

16,5 % total system efficiency (17,2 % effective system

efficiency) may be attainable if the cells are actively cooled

so as to remain at a fixed temperature of 60 °C. Here,

effective system efficiency was defined as the total annual

A.C. energy output divided by the annual global irradi-

ance that would be received by a static system of similar

aperture area, irrespective of the type of PV system that

was being discussed (i.e. one-axis, two-axis, or concen-

trator).

R egarding the prospects for a concentrated VLS-PV

plant, it was concluded that a single 1 m2 concentrator PV

module exposed to a light flux of 400 suns (where 1 sun is

Table B.3 Induced impacts of transferring a PV module manufacturing facility

Total induced Value added Job creation

(1 000 USD) (1 000 USD) (man-years)

I L I L I L

1 Photovoltaic 16 294 16 294 1 485 1 485 87 87

6 Misc. petroleum and coal products 205 1 248 160 975 23 138

7 Petroleum refineries 271 1 612 53 315 8 45

8 Electricity and water 335 1 492 85 377 12 53

18 Metals smelt ing and pressing 323 4 477 113 1 573 16 223

19 Fabricated metal products 301 1 082 113 406 16 58

20 Non-electrical machinery 44 599 18 248 3 35

22 Electrical and electronic equipment 501 16 350 170 5 567 24 791

24 Chemicals 1 039 3 718 318 1 138 45 162

25 Rubber and plastics 656 1 210 217 400 31 57

28 Commerce and transport 195 1 244 111 709 21 132

31 Insurance 500 539 157 169 28 30

32 Other services 648 2 326 415 1 492 73 262

Others 947 5 389 446 2 317 103 497

Total 22 259 57 581 3 862 17 099 489 2 570Induced production coeff icient 1,36 3,53

Figure B.12 The PETAL (Photon Energy Transformation and

Astrophysics Laboratory) 400 m2 aperture parabolic dish

reflector at Sede Boqer (the square screen at its focal point is

1 m x 1 m in size)

Executive Summary B

20

here defined as 1 000 W/ m2) would produce nearly 100 kW

of electric power. Table B.4 compares a number of area-

related output parameters of interest for all four types of

systems.

The estimation of the cost of a 30 MW turnkey project

was approximately 136 500 USD per concentrator (CPV)

unit, where in a typical meteorological year each unit would

generate 154 000 kWh of A.C. electricity at Sede Boqer.

When assuming 30-year financing, the electricity would

work out to 0,045 USD/ kWh in the case of 3 % interest

Table B.4 Comparison of predicted area-related performance parameters for various VLS-PV systems at Sede Boqer

Static 30° t ilt One-axis tracking Two-axis tracking Concentrator (CPV)

System yield (kWh ⋅ kW -1 ⋅ y-1) 1 644a 2 071a 2 279a 1 754b

Yield per PV module area (kWh ⋅ m -2 ⋅ y-1) 189 238 262 154 000

Yield per 400 m2 light capture area (kWh⋅y-1) 75 600 95 200 104 800 154 000

Land area/light capture area 2,89 3,42c 7,90d 7,90d

Yield per land area (kWh ⋅ m -2 ⋅ y-1) 65,4 69,6 33,2 48,7

aSolarex MSX64 modules.bModified SunPower Heda 303 cells.cRancho Seco plant.dHesperia Lugo plant.

and to 0,064 USD/ kWh in the case of 6 % interest. By

considering annual O&M costs to be 2 % of the capital

cost, 0,018 USD/ kWh would be added to the electricity

costs given above.

There is a wealth of hardware and performance data

available for non-concentrator PV systems for validation,

while the experimental situation is sparse for concentrated

PV systems. However, it might be concluded that the

economics of a concentrator VLS-PV plant could turn out

to be attractive.

21

Scenario Studies and Recommendations

EXECUTIVE SUMMARY C

Scenario studies and

recommendations

Through Parts I and II, it was suggested that the VLS-PV

system would be an attractive energy source in the 21st

century. However, the long-term sustainable operation of

VLS-PV systems must be discussed considering local ben-

efits to be brought about by introduction and expansion

of the VLS-PV system. Then the step-by-step enlargement

of the PV system might be an effective way to prevent

financial, technological and environmental risks caused by

its rapid development. In addition, it is expected that this

activity will be continued with properly allotted efforts for

further quantitative discussion and necessary evaluation of

VLS-PV to improve our knowledge, to overcome its

defects and to establish an international network with other

interested people. In Part III, based on the generalized

understandings from Parts I and II, scenario studies were

carried out and ‘recommendations’ to stakeholders as

conclusions were described.

In the scenario studies, a concept for the sustainable

growth of VLS-PV was discussed, which included both

the long-term economic aspects and the life-cycle point

of view, and a development scenario assuming actual stages

was proposed. Further, the financial and organizational

sustainability on proposed stage was discussed.

Finally, in order to propose mid- and long- term

scenario options that would enable the feasibility of

VLS-PV, recommendations for considerable stakeholders

were given to realize such long-term targets gradually.

C.1 SUSTAINABLE GROWTH OF THEVLS-PV SYSTEM CONCEPT

VLS-PV has a huge generating power capacity and it

will be more feasible to enhance the capacity gradually.

Therefore, for the introduction of VLS-PV, a develop-

ment scheme for sustainable growth is needed. Considering

the economic aspects of VLS-PV is also important for a

sustainable initiative.

The domestic production of PV modules is one of the

most important issues for the successful introduction of

VLS-PV. As an example, a conceptual scheme of PV

module manufacture through technology transfer to the

region, as shown in Figure C.1, was considered. PV

modules produced at the facilities will be installed in desert

areas as part of a centralized system.

The power-generating capacity of the system will be

reinforced yearly with operating PV module facilities and

a VLS-PV system will be developed. Figure C.2 shows the

sustainable scheme for VLS-PV development, which

supposes that the capacity of VLS-PV will be over 1 GW

in 28 years, and will reach 1,5 GW in 43 years. With the

replacement of PV modules and facilities, and with

operating facilities, the VLS-PV will be a sustainable power

plant. Further, when a variety of VLS-PV components are

produced near the VLS-PV and waste management, like

recycling, is introduced, ‘scrap and build’ for VLS-PV will

be realized, as shown in Figure C.3.

In the economic analysis, the generation cost for VLS-

PV was estimated to be around 3–5 US cents/ kWh for

1,5 GW. However, to obtain long-term economic benefits,

providing advantageous incentives will be needed in the

first stage. Further, for deploying domestic/ regional manu-

facturing of VLS-PV system components, it is necessary

to build a financial mechanism for investing in facility

operation, not only investment for constructing the facility.

Some organizational/ institutional support will be indispen-

sable and this will contribute not only to profitability in the

long term but also to successful technology transfer.

Figure C.1 Conceptual view of a sustainable, technology

transfer scheme

Executive Summary C

22

C.2 POSSIBLE APPROACHES FOR THE FUTURE

The VLS-PV scheme will be a project that has not been

experienced before. Therefore, to realize VLS-PV it is

necessary to identify issues that should be solved and to

discuss a practical development scenario. Focusing on the

technical development, a basic concept for VLS-PV devel-

opment, which consisted of the following four stages, was

proposed. Although the capacity of the first PV system

was set at 25 MW, R&D stage (S-0) was assumed as the

first stage to verify the basic characteristics of the PV sys-

tem.

• S-0: R &D stage (four years)

• PV system: 5 × 500 kW research system

• Price of PV module: 4 USD/ W

Figure C.2 Sustainable scheme for VLS-PV development

Figure C.3 Example of a concept for sustainable growth of VLS-PV

23

Scenario Studies and Recommendations

• PV module: import from overseas

• Inverter: import from overseas

• S-1: Pilot stage (three years)

• PV system: 25 MW pilot system

• Price of PV module: 3 USD/ W

• PV module: import from overseas

• Inverter: import from overseas

• S-2: Demonstration stage (three years)

• PV system: 100 MW large-scale system

• Price of PV module: 2 USD/ W

• PV module: domestic/ regional production

• Inverter: import from overseas

• S-3: Deployment stage (five years)

• PV system: around 1 GW VLS-PV system with

energy network

• Price of PV module: 1 USD/ W

• PV module: domestic/ regional production

• Inverter: domestic/ regional production

To develop VLS-PV, there are many technical and non-

technical aspects that should be considered in each stage.

These are summarized in Table C.1.

S-0: R&D stage (four years)

Five 500 kW PV systems will be constructed and operated

to verify the basic characteristics of the PV system in a

desert area. The reliability of VLS-PV in a desert area and

the requirements for grid connection will be mainly

examined and investigated as technical issues. Conditions

for site selection, planning of co-operation frameworks,

including training engineers, and funding schemes will be

investigated as non-technical issues.

S-1: Pilot stage (three years)

A 25 MW PV system will be constructed and operated to

evaluate and verify preliminary characteristics of a large-

scale PV system. Technical issues here are of a higher level

and shift to concentrated and grid-connected PV systems,

for building VLS-PV technical standards. PV module

facilities will be constructed and operated although the

PV modules constructed would be imported. PV module

production will be introduced in the next stage. Further,

development for preventing desertification, such as

vegetation and plantations, will also be started at this stage.

S-2: Demonstrat ion stage (three years)

A 100 MW PV system will be constructed and operated

to research methods of grid-connected operation and

maintenance when VLS-PV actually takes on part of

the local power supply. The knowledge, on connecting

VLS-PV to the existing grid line, obtained at this stage

will be technical standards for deploying VLS-PV. Non-

technical issues will also advance for industrialization. Mass

production of PV modules will be carried out and BOS

production on-site will be started towards a deployment

stage.

S-3: Deployment stage (f ive years)

A 1 GW PV system will be operated to verify the capability

of VLS-PV as a power source. Although the technologies

for generating and supplying electricity will be nearly

completed, for deployment of VLS-PV in the future, some

options such as demand control, electricity storage and

component recycling will be required. These will contrib-

ute to building the concept of the solar breeder, which is

similar to ‘scrap and build’ shown in Figure C.3, and some

business plans for VLS-PV will be proposed.

The implementation of VLS-PV will activate the world PV

industries in a wide range of technologies involving a vast

range from solar-cell production to system construction.

However, to achieve the final stage, the previous stages

from S-0 to S-2 are important, and the developments in

each stage should be carried out steadily.

Table C.1 Summary of the VLS-PV development scenario

Sub-stage

Technical issues Non-technical issues

S-0: Scale of system: 5 x 500 MW • Examination of the reliability of • Site selection for VLS-PV based on

R&D stage Module cost: 4 USD/W a PV system in a desert area various condit ions

Module: import from overseas • Examination of the required • Project planning, including training

Inverter: import from overseas ability of a PV system for grid engineers and funding

connection

S-1: Scale of system: 25 MW • Development of the methods of • Development of the area around

pilot stage Module cost: 3 USD/W O&M for VLS-PV VLS-PV to prevent desert if ication

Module: import from overseas • Examination of the control of • Training engineers for PV module

Inverter: import from overseas power supply from a PV system production on-site

to grid line

S-2: Scale of system: 100 MW • Development of the technical • Training engineers for mass production

demonstration stage Module cost: 2 USD/W standards for O&M of VLS-PV, of PV modules and for BOS production

Module: domestic/regional including grid connection on-site

production • Preparation for industrialization by

Inverter: import from overseas private investment

S-3: Scale of system: 1 GW • Building the concept of ‘solar breeder’ from the viewpoint of technical and

deployment stage Module cost: 1 USD/W non-technical issues

Module: domestic/regional

production

Inverter: domestic/regional

production

Executive Summary C

24

C.3 FINANCIAL AND ORGANIZATIONALSUSTAINABILITY

The proposed scenario for the development and introduc-

tion of very large-scale PV systems in deserts has basically

four stages: R &D, pilot, demonstration and deployment

(commercial) stages. The characteristics of these stages range

from theoretical to technical, i.e. through the techno-

economic, socio-economic and purely commercial

characteristics. Each stage of growth has its own charac-

teristics and cost structure.

From a viewpoint of the financial scenario, the feasibil-

ity stage should be settled before the R &D stage. The

estimated investment levels for these phases range from

1–2 MUSD for the feasibility stage to 4 000 MUSD for

the commercial stage.

An indicative portfolio of funding sources for each stage

is identified. Since the first two stages are not supposed

to be commercial, but are theoretical and experimental,

no further exploitation or return on investment (ROI)

calculation is undertaken. At this moment, the costs and

funding arrangements of the commercial stage can only

be estimated. The third stage (the demonstration stage)

should generate cost and income details for a full invest-

ment proposal for such a large plant.

There are six potential sources of financing the invest-

ment. In Table C.2, a first estimate of possible contribu-

tions from these sources is given:

• Direct subsidies may be provided for demonstration,

experimental, export promotion, or development

co-operation reasons by governmental bodies. The closer

a given situation is to being commercial, the lower the

direct subsidies that are expected to be granted.

• The provision of soft loans and green money may be

dr iven by similar motives, but they can also be

provided by private capital. The loan money should

increase as the project becomes more commercial to

leverage the ROI for the shareholders.

• Equity and in-kind contributions are investments by

the shareholders of the project. Examples of in-kind

contributions are office housing and free management

services.

• Import duties may be exempted or reduced to stimulate

the uptake of new or renewable energy technologies in

a particular country.

• Green certificate buy-off may constitute an up-front

contribution of a utility that subsidizes the project and

receives green certificates in return.

• Other instances may be profit tax advantages of inves-

tors regarding the project.

Basically, all net investment costs, interest and profits should

be recovered through exploitation of the plant. The net

investment is the investment after subtracting subsidies and

fiscal advantages. There are four recognized sources of

income during the exploitation stage:

• electricity power sales

• opportunity costs, to be explained later

• green certificates, not to be double-counted with a

possible up-front investment

• tax incentives, such as reduced VAT and exemption

from pollution taxes.

In Table C.3, an overview of the income and recurring

costs from the exploitation is given.

The feasibility stage is easily supported when applying

to national or multinational governmental organizations.

The size and novelty of the project is expected to gear

sufficient interest to achieve the involvement of the World

Bank, the EU, and the respective national governments of

interested OECD countries. There are no exploitation costs

or income at this stage. At a cost of around 30–40 MUSD,

this R &D/ pilot project would be a project of a relatively

large size for individual governments to bear, but not for

two or more governments or multilateral organizations. As

Table C.2 Example of contribution towards investment by various sources of co-funding in the different stages of the

introduction of VLS-PV

Source Direct subsidies (Soft) loans Equity and Import duty Green cert if icate Others, such Totala

including green in-kind reduction buy-off and as profit tax

Stage money JI funds advantages

1. Feasibility 75 % 0 % 15 % – – 10 % 100 %

2. R&D/pilot 50 % 30 % 10 % 10 % – – 100 %

3. Demonstration 25 % 30 % 15 % 10 % 15 % 10 % >100 %

4. Deployment/

commercial 5 % 65 % 15 % 10 % 10 % 10 % >100 %

aRedundancy of funding is necessary to reduce risks.

Table C.3 Estimated income as a percentage of total recurring costs after subsidy of the different stages of the introduction

of VLS-PV

Source of funding Electricity Opportunity Green Tax Others Total

Stage costs cert if icates incentives

1. Feasibility – – – – – –

2. R&D/pilot 50 % – 30 % 20 % – 100 %

3. Demonstration 30–50 % 0–17 % 18 % 0–30 % – 48–115 %

4. Deployment/

commercial 50 % 10 % 20 % 20 % – 100 %

25

Scenario Studies and Recommendations

it has interesting technical and social aspects, there will be

multiple instruments to apply for funding. The strongly

reduced investment by heavy subsidies will enable the

electrical power to be sold favourably. Thus electricity

accounts for 50 % in the income of the plant. Other sources

of income are green certificate values and tax incentives.

The subsidy for the cost and financing of a 100 MW

demonstration plant will be reduced. The exploitation could

be economically attractive, depending on several assump-

tions. The first and most important are the value of green

certificates and the value of so-called opportunity costs.

These opportunity costs are costs avoided by certain

companies or increased income for such companies, due

to the existence of the VLS-PV project. The deployment/

commercial stage will have to benefit from reduced costs

of PV modules and increased cost of power and of green

certificates. However, it may be too early to give more

details regarding such a 1 GW plant.

It is recommended to make a full-scale feasibility study

for an R &D project and a 100 MW demonstration plant.

This feasibility study should identify targets and location,

and fully secure funding sources and electricity outlets for

both stages. Without funding identified and secured for

the 100 MW demonstration plant, the R &D stage should

not be implemented.

C.4 RECOM M ENDATIONS

As an overall conclusion of this work, recommendations

are given here to realize long-term targets based upon the

results of the studies performed in the IEA Task VIII. The

adoption of VLS-PV will require four steps, as shown in

Figure C.4:

1. Absorbing the concept of VLS-PV in a desert environ-

ment

2. Considering a long-term scenario approach

3. Positioning yourself in terms of your strategy

4. Making an action plan and allocating resources

In this report, concrete information on VLS-PV is given,

so the first two steps can be taken into consideration.

The generalized understandings and recommendations

on a policy level give direction to those who consider the

adoption process. To support those willing to consider the

third and fourth steps, an initial checklist is given.

C.4.1 General understandings

Based on this report, the considered stakeholders may

recognize the following valuable findings.

VLS-PV can contribute substantially to global energy needs

• The world’s deserts are so large that covering 50 % of

them with PV units would generate 18 times the

primary energy supply of 1995.

• All global energy scenarios project solar PV energy to

develop into a multi-gigawatt energy generation

option in the first half of this century.

VLS-PV can become economically and technologically

feasible

• Electricity generation costs are between 0,09 and 0,11

USD/ kWh, depending mainly on annual irradiation

levels (with module price 2 USD/ W, interest rate 3 %,

salvage value rate 10 %, depreciation period 30 years).

These costs can come down by a factor of 2–3 by the

year 2010.

• The PV technology is maturing with increasing

conversion efficiencies and decreasing prices per watt;

projected prices of 1,5 USD/ W around the year 2010

would enable profitable investment and operation for a

100 MW PV plant.

• Solar irradiation databases now contain detailed infor-

mation on irradiation in most of the world’s deserts.

VLS-PV can contribute considerably to the environment

• The life-cycle CO2 emission is as low as 13 g-C/ kWh;

this is mainly due to the module production and the

array support. This should be compared with the value

of 200 g-C/ kWh for just the fuel component of

conventional fossil-burning power plants.

• The environmental issues for which VLS-PV may

provide a solution are global warming, regional

desertification and local land degradation.

VLS-PV can contribute considerably to socio-economic

development

• Plant layouts and introduction scenarios are available

in preliminary versions. I/ O analysis concluded that

25 000–30 000 man-years of local jobs for PV module

production will be created per 1 km2 of VLS-PV

installed.

VLS-PV development needs a long-term view and

consistent policy

• To reach the level of a 1 GW system, four intermediate

stages are recommended: R &D stage, pilot stage, dem-

Figure C.4 Flowchart for recommendation

Executive Summary C

26

onstration stage, and deployment (commercial) stage.

From stage to stage, the system scale will rise from

2,5 MW to 1 GW, the module and system cost will go

down by a factor of 4, and manufacturing will be shifted

more and more to the local economy.

• In the concept of sustainable local economic growth,

the first local PV module manufacturing facility has an

annual output of 5 MW. This local production provides

for construction of the local VLS-PV system. In subse-

quent years, four more 5 MW module manufacturing

facilities are brought into operation, so that annually

25 MW is supplied to the local VLS-PV system. After

10–15 years, a module production facility of 50 MW is

put into operation. Every 10 years this facility will be

replaced by a more modernized one. Thus after

approximately 40 years a 1,5 GW VLS-PV power

station will be in operation, and the local manufactur-

ing facility will supply for replacement. In this way,

local employment, and thus the economy, will grow

sustainably.

• To realize the final commercial stage, a view to financ-

ing distribution has been developed for all of the three

previous stages, consisting of direct subsidies, soft loans,

equity, duty reduction, green certificates and tax

advantages. It is clear that direct subsidies will play an

important role in the first stage.

C.4.2 Recommendations on a policy level

From the global energy situation, global warming and

other environmental issues, as well as from the case studies

and scenarios, it can be concluded that VLS-PV systems

will have a positive impact. To secure that contribution, a

long-term scenario (10–15 years) on technological,

organizational and financial issues will be necessary.

Action now is necessary to unveil the giant potential of

VLS-PV in deserts. In such action, involvement of many

actors is welcome. In particular, the following are recom-

mended on a policy level:

• National governments and multinational institutions

adopt VLS-PV in desert areas as a viable energy

generation option in global, regional and local energy

scenarios.

• The IEA PVPS community continues Task VIII for

expanding the study, refining the R &D and pilot stages,

involving the participation by desert experts and

financial experts, and collecting further feedback infor-

mation from existing PV plants.

• Multilateral and national governments of industrialized

countries provide financing to generate feasibility studies

in many desert areas around the world and to imple-

ment the pilot and demonstration phases.

• Desert-bound countries (re-)evaluate their deserts not

as potential problem areas but as vast and profitable

(future) resources for sustainable energy production. The

positive influence on local economic growth, regional

anti-desertification and global warming should be

recognized.

C.4.3 Checklist for specific stakeholders

To decision-makers in industrialized countries

You obviously have a long-term view of the world energy

market trends and the need to provide a national energy

outlook.

• Have you considered the future possibility of VLS-PV

for your industries, which may become major enter-

prises controlling the world energy market?

• Do you have a step-by-step plan for R &D to make

good use of the extensive capabilities in photovoltaic

technology when the world energy problem arrives?

• Do you have a view to initiate, continue and extend

bi- or multilateral international collaboration with

those developing countries which have abundant solar

energy?

• Do you have funds available for R &D or pilot

programmes with training courses to introduce PV tech-

nology into developing regions, especially around deserts

as a first stage of a consistent step-by-step approach?

• Do you have strategies in place to maintain regional

sustainability and to consider a moderate technology

transfer scenario when planning the further develop-

ment of developing countries?

• Have you considered using your influence to mobilize

multilateral institutions to stimulate VLS-PV?

To decision-makers in developing countries

You obviously are aware of the coming world energy

problem in 20–30 years.

• Did you include solar PV energy as one of the most

favourable renewable energy options when national

master plans for energy supplies were discussed?

• Have you considered the opportunity that your country

will be able to export PV energy to neighbouring

regions and that new jobs will be brought to your

people?

• Are you aware of the fact that PV technology has

already proven itself to be a cost-competitive energy

source for rural electrification and is still being improved

very rapidly? In particular, are you aware that it is

especially effective for stabilizing rural lives?

• Have you considered a regional development plan that

utilizes abundant electricity production and vast lands?

• Have you settled on a step-by-step, long-term approach

that starts with solar home systems or mini-grids as the

first stage and finally reaches VLS-PV in 20–30 years?

• Do you have a plan to cultivate and gradually raise a

domestic PV specialists’ society from an early stage to a

developed stage?

• Have you already asked for support from the variety of

financial institutions you can utilize?

To decision-makers in oil-export ing countries

You obviously are aware of the fact that many oil-export-

ing countries around desert areas also have an everlasting

natural resource: solar energy.

27

Scenario Studies and Recommendations

• Are you aware that you can export PV energy to neigh-

bouring regions as well?

• Do you know that PV technology has already been

proven as a cost-competitive energy source for rural

electrification and is still being improved rapidly?

• Did you develop a long-term view of the future world

energy market and your strategy including the new level

of photovoltaic power plants and industries? Are you

aware that it will bring you opportunities for high-tech

industries and new jobs?

• Can you confirm the study results that a 100 MW PV

power plant will be economically attractive in an oil-

exporting country? Have you discovered good condi-

tions in interest rates, the value of green certificates and

the value of opportunity benefits from oil savings?

• Have you decided to invest in the development of the

world photovoltaic business?

• Did you choose an appropriate scale for starting

towards VLS-PV?

To f inancing institut ions and banks

You are presumably aware of the fact that the market

potential for VLS-PV amounts to 2 billion people world-

wide.

• Are you aware of the Task VIII study results that show

the size of indicative investment levels for the 100 MW

demonstration stage and the 1 GW deployment stage

corresponds to 500 and 4 000 MUSD respectively? Do

you consider this much different from the magnitude of

hydropower or infrastructure projects in a budgetary

sense?

• Do you respect the following funding scenario study

for a 100 MW demonstration stage and a 1 GW

deployment stage? They are economically attractive in

some cases, assuming that 30–65 % of total investment is

met by soft loans with 4 % interest. Another portion is

expected to come partially from subsidy, equity, tax

reduction, etc.

• Can you positively support a full-scale feasibility study

for a pilot project and for a 100 MW demonstration

plant as a continuation of the Task VIII? This will iden-

tify targets and locations and will secure the funding

sources and electricity outlets for both stages.

• Can you support the pilot stage and the 100 MW

demonstration stage according to the results of this

study?

• Could you consider a low-interest soft loan on a long-

term basis for the initiation of VLS-PV system projects

around desert areas?

To PV industry associat ions and mult inational industries

You are obviously aware of possibilities for future market

growth in southern countries.

• Are you aware of the future possibility of VLS-PV for

PV industries? They may become major enterprises

controlling the world energy market.

• Are you confident that the photovoltaic technology and

market will become competitive on a worldwide level

within 20 years?

• Can you ensure that the prices for solar PV energy will

be reduced by a factor of a half to a quarter within the

next decade?

• Can you support and invest in local industries to take

off according to the technology transfer scenario?

To PV specialists and the academic community

You know that fundamental research will generate new

seed technologies for VLS-PV.

• Can you confirm expected directions such as very high-

efficiency PV cells, high-concentration optics, organic

polymer PV cells, chemical energy transportation

media like hydrogen or methanol, superconducting

power transmission and so on?

• Did you formulate and assist a PV specialist society in

developing countries in co-operation with top leaders

in those countries?

• Will you join our continuing work, seeking the realiza-

tion of VLS-PV systems? Expected work items may

include more precise case studies for specific sites

and funding, proposals for R &D co-operation plans,

other possibilities in technological variety, resource

evaluation, additional value analysis and so on.

To power ut ilit ies

You have clearly recognized that the world energy

market structure will change very drastically in the near

future.

• Can you confirm business opportunities in photovoltaics

within the next decades?

• Can you confirm that a power transmission scenario is

possible according to our study results? Additional tie-

line construction of less than 100 km, for connecting

VLS-PV through existing national power grids to a load

centre, will raise the electricity price by less than 1 US

cent/ kWh. One example is a transmission operation in

co-operation with coal-burning power stations located

on a colliery.

• Are you ready to invest in photovoltaic industries and

foster technological societies with a long-term view for

the future world energy market?

To the Internat ional Energy Agency

You are clearly aware that the diversification of energy

resources and the development of alternative energy are

essential for overcoming the world energy problems within

the next decades.

• Can you confirm our view that solar PV energy is one

of the most favourable options for future electricity

production?

• Can you confirm your continuous support of the IEA

PVPS Implementing Agreement on the basis of the long-

term world energy outlook?

Executive Summary C

28

• Would you support our idea about multilateral activities

between IEA member countries and developing countries?

• Can you organize the higher level of IEA PVPS activi-

ties including demonstration projects for VLS-PV?

• Do you want to support a full-scale feasibility study

corresponding to a pilot project and a 100 MW dem-

onstration plant? This will identify targets and location,

and fully secure funding sources and electricity outlets

for both stages.

• Can you support and enhance the continuing work in

the IEA PVPS Task VIII? Expected work items are to be:

• more precise case studies for specific sites including

detailed local conditions and funding sources as well

as demand application,

• proposals for the first or second stage of co-operation

plans to be submitted to financial institutions,

• comprehensive evaluation of other possibilities of a

technological variety such as tracking, concentrator

and advanced PV cells,

• resource evaluation of VLS-PV by means of remote

sensing technology,

• investigation of additional effects of VLS-PV on the

global environment such as global warming and

desertification,

• expansion of evaluating approaches to other types of

PV mass applications in the 21st century, including

value analysis in the economy, environment, socio-

economy and others.

The world’s deserts are sufficiently large that, in theory, covering a fraction of their landmass with PV systems could

generate many times the current primary global energy supply. Moreover, the energy produced is from solar radiation

– a clean and renewable source – hence such systems would have the potential to contribute massively to the

protection of the global environment.

Energy from the Desert is an extensive and high-level international study, representing the accumulated research of

the world experts involved in Task VIII of the IEA PVPS Programme. This Summary highlights the principal findings of

the full report (available separately).

To date, the market focus for photovoltaics has been on small to medium, stand-alone or building-integrated power

systems, which have proven, but as yet not realized, the great potential of this technology. Energy from the Desert

evaluates the feasibility, potential and global benefits of very large scale photovoltaic power generation (VLS-PV)

systems deployed in desert areas and each generating from 10MW to several gigawatts.

The study details the background and concept of VLS-PV, maps out a development path towards the realization of

VLS-PV systems, and provides firm recommendations to achieve long-term targets, based on the findings of the IEA

Task VIII experts. Critical aspects examined are:

• photovoltaic technologies, systems design and plant operation

• finance, cost-benefits and profitability

• impact on and benefit to global, regional and local environment

• policy-level and investment issues

Energy from the Desert is the first study to provide a concrete set of answers to the questions that must be addressed

in order to secure and exploit the potential for VLS-PV technology and its global benefits. It will be invaluable to

government, energy planners, policy makers, utilities and international organizations assessing the potential for this

technology, PV systems manufacturers and infrastructure providers wishing to develop this new market and

consultants, scientists, researchers and engineers involved in the field.

THE EDITOR

Energy from the Desert is edited by Professor Kosuke Kurokawa, Operating Agent-Alternate, IEA Task VIII with

contributions from a panel of fourteen acknowledged international experts from eight countries. Professor Kurokawa

has 30 years’ experience in energy systems technology including HVDC transmission and solar photovoltaics. He is

currently engaged in researching a wide variety of photovoltaics systems concepts from small AC modules to very

large scale PV systems at the Tokyo University of Agriculture and Technology (TUAT).

Energy from the Desert

James & James (Science Publishers) Ltd

8-12 Camden High Street, London NW1 0JH, UK

www.jxj.com